[PATCH] D22460: [X86][AVX] Added support for lowering to VBROADCASTF128/VBROADCASTI128

[Elena Demikhovsky via llvm-commits]

delena added inline comments.

================
Comment at: test/CodeGen/X86/vector-shuffle-256-v4.ll:1365
@@ -1366,4 +1364,3 @@
 ; AVX512VL:       # BB#0:
-; AVX512VL-NEXT:    vmovdqa64 (%rdi), %xmm0
-; AVX512VL-NEXT:    vinserti32x4 $1, %xmm0, %ymm0, %ymm0
+; AVX512VL-NEXT:    vbroadcasti128 {{.*#+}} ymm0 = mem[0,1,0,1]
 ; AVX512VL-NEXT:    retq
----------------
The selected instruction here is from AVX2 set. The same patterns should be added to AVX-512.


Repository:
  rL LLVM

https://reviews.llvm.org/D22460